Spitzer/MIPS 24μm observations of HD209458b: Three eclipses, two and a half transits, and a phase curve corrupted by instrumental sensitivity variations

We report the results of an analysis of all Spitzer/MIPS 24 μm observations of HD209458b, one of the touchstone objects in the study of irradiated giant planet atmospheres. Altogether, we analyze two and a half transits, three eclipses, and a 58hr near-continuous observation designed to detect the planet's thermal phase curve. The results of our analysis are: (1) a mean transit depth of 1.484% ± 0.033%, consistent with previous measurements and showing no evidence of variability in transit depth at the 3% level. (2) A mean eclipse depth of 0.338% ± 0.026%, somewhat higher than that previously reported for this system; this new value brings observations into better agreement with models. From this eclipse depth we estimate an average dayside brightness temperature of 1320 ± 80 K; the dayside flux shows no evidence of variability at the 12% level. (3) Eclipses in the system occur 32 ± 129 s earlier than would be expected from a circular orbit, which constrains the orbital quantity ecos ω to be 0.00004 ± 0.00033. This result is fully consistent with a circular orbit and sets an upper limit of 140ms^-1 (3σ) on any eccentricity-induced velocity offset during transit. The phase curve observations (including one of the transits) exhibit an anomalous trend similar to the detector ramp seen in previous Spitzer/IRAC observations; by modeling this ramp we recover the system parameters for this transit. The long-duration photometry which follows the ramp and transit exhibits a gradual 0.2% decrease in flux over 30hr. This effect is similar to that seen in pre-launch calibration data taken with the 24 μm array and is better fit by an instrumental model than a model invoking planetary emission. The large uncertainties associated with this poorly understood, likely instrumental effect prevent us from usefully constraining the planet's thermal phase curve. Our observations highlight the need for a thorough understanding of detector-related instrumental effects on long timescales when making the high-precision mid-infrared measurements planned for future missions such as EChO, SPICA, and the James Webb Space Telescope.